How do you ensure data consistency in a project? Have there been cases of discrepancies?

To ensure data consistency on the project, transactions and locking mechanisms were used to prevent race conditions and partial updates. In distributed systems, patterns such as two-phase commit (2PC) or the use of consensus logs were applied.

Data integrity checks at the database level (e.g., foreign key constraints, uniqueness) and validation at the application level were also employed.

There were cases of discrepancies, for example, due to network failures or processing errors, which led to data desynchronization. Monitoring, auditing, periodic data reconciliation, and mechanisms for compensating transactions or retries were used to detect and correct these issues.